The present invention relates to a method of measuring electromagnetic radiation that is radiated from a surface of an object that is irradiated by electromagnetic radiation given off by at least one radiation source, whereby the radiation given off by the radiation source is determined by at least one first detector, and the radiation given off by the irradiated object is determined by at least one second detector that measures the radiation.
A method of this type is known, for example, from U.S. Pat. No. 5,490,728 A in conjunction with the manufacture of semiconductor substrates in a reaction chamber. In this case, the electromagnetic radiation given off by the radiation source is naturally superimposed with a waviness that is undesired and occurs due to fluctuations of the line voltage or due to phase controls. Unfortunately, no influence can be had on this waviness, and can also not be intentionally selected. The waviness is therefore only suitable, if at all, to a limited extent for an intentional utilization as a characteristic of the radiation given off by the radiation source.
Reference is also made to DE-A-26 27 753, which discloses an apparatus for measuring and controlling the thickness of optically effective thin layers during build up thereof in vacuum coating units. The measurement and control is achieved by detecting the reflection or transmission characteristic of layer thicknesses between fractions of and multiples of the essentially monochromatic measurement light that is utilized, and by interruption of the coating process when a predetermined layer thickness has been achieved. The apparatus comprises a measurement light source for a focused measurement light beam, a chopper, a beam splitter that is disposed in the axis of the measurement light beam at an angle of 45xc2x0, a measurement light receiver that is connected in series with a monochromatic illuminator, as well as a differentiator for the measurement signal and an interrupter for the coating process. Furthermore, DE-A-42 24 435 discloses an optical interface for the infrared monitoring of transparent disks, whereby the light of an infrared radiation source is conducted by a beam wave guide into the interior of the interface where it is emitted for exposing the surface of the disk. The radiation reflected at the disk that is to be monitored is received by the input of another beam wave guide and is conveyed by the beam wave guide via a daylight filter to a photo detector. U.S. Pat. No. 5,270,222 furthermore discloses a method and an apparatus for a diagnosis and prognosis during the manufacture of semiconductor devices. The apparatus has a sensor for the diagnosis and prognosis that measures various optical characteristics of a semiconductor wafer. The sensor has a sensor arm and an optoelectronic control box for conducting coherent electromagnetic or optical energy in the direction of the semiconductor wafer.
It is therefore an object of the present invention to provide a method of the aforementioned general type with which the measurement of electromagnetic radiation, and the determination of the pyrometers and values drived therefrom, can be carried out in a straightforward manner and more precisely.
The stated object is inventively realized in that the radiation given off from at least one radiation source is actively modulated with at least one characteristic pyrometer, and in that the radiation determined by the second detector is corrected by the radiation determined by the first detector to compensate for the radiation of the radiation source reflected from the object. The radiation source is, preferably a heat lamp and the irradiated object is preferably a semiconductor substrate that is subjected to a thermal treatment.
Due to the intentional, active and hence known modulation of the radiation source with a characteristic pyrometer it is possible to more precisely differentiate the difference between the radiation radiated from the object itself, and necessary for the determination of the characteristics of the object, from the radiation of the radiation source reflected from the object. In this way, it is possible to determine more precisely and in real time the characteristics of the object, for example the temperature, the emissivity, the transivity, the reflectivity, or the layer thicknesses or characteristics of a material that is on the object and differs from the material of the object.
Pursuant to one particularly advantageous embodiment of the invention, the active modulation of the radiation given off by the radiation source for the characterization thereof, is used during the correction of the radiation determined by the second detector. Due to the active and hence known modulation of the radiation given off by the radiation source, the characterization and hence differentiation of this radiation from the actually to be measured radiation that is given off by the object is particularly simple, reliable and quantitatively accurate.
The radiation given off by the radiation source is preferably modulated with respect to amplitude, frequency and/or phase. Depending upon the existing conditions and requirements, the type of modulation can be selected as desired, whereby the type of modulation can be selected in particular also with respect to the simplicity and reliability of the modulation process, but also of the evaluation process and of the detection process. In this connection, amplitude modulation means the modulation of the modulation amplitude. However, the process preferably involves intensity modulation, the amplitude of which is not modulated, but rather possibly slowly varied.
In addition to the type of modulation, it is also possible to utilize every signal shape of the modulation. However, particularly advantageous, during an amplitude modulation, is the use of a signal shape having a signal pattern that is as continuous as possible. This has the advantage that also during a Fourrier transformation high frequencies essentially do not occur and therefore the number of scans per unit of time during the detection or processing of the detected signal can remain low, so even with a simple evaluation process a good and accurate measurement can be carried out.
In general, the modulation of the characteristic pyrometer can be effected with a periodic or non periodic signal. A non periodic modulation can be obtained, for example, in that the characteristic pyrometer is linked with a positive or negative increment which is generated by means of a random mechanism, via a linking operation (e.g. addition, multiplication or a linking with a look-up-table). In this connection, after a certain interval of time has elapsed the increment is respectively predetermined pursuant to a random principle. The time interval itself can in this connection be constantly determined pursuant to a predefined function or again pursuant to a random principle. The important thing with the non periodic modulation is that the pyrometer (increment and/or time interval) determined by random principles be known and be available within an evaluation device or an evaluation process for signal analysis. The pyrometers (increment and/or time interval) determined by a random principle can satisfy an arbitrarily predefined distribution function. They can, for example, be distributed uniformly, in a Gaussian fashion or pursuant to a Poisson distribution, as a result of which the respective expected values of the pyrometers are similarly predefined. The advantage of a non periodic modulation is that as a result periodic disruptive influences can be suppressed.
A further advantageous embodiment of the invention consists in that the radiation source comprises a plurality of individual radiation sources, for example a plurality of lamps, that can be combined into one or more lamp banks. Pursuant to advantageous embodiments in conjunction with radiation sources that comprise a plurality of lamps, the radiation of at least one of the lamps is modulated. Although the modulation of the radiation of one lamp can be adequate to achieve the advantages of the inventive method, the modulation of only one lamp in general delivers a practical result only by limiting the universality of the measuring process. A particularly straight forward control of the lamps with a single power switch is also provided in particular if the radiation of at least two lamps or of all lamps are modulated in the same manner. Advantageously the radiation of only one or some of the lamps is modulated in order to avoid undesired reflections.
Depending upon the applications and conditions, it is, however, also advantageous to modulate the radiation of a lamp differently, for example if the lamp radiation is to be differentiated as a function of the position of the lamps or from the respectively specific lamp relative to the radiation of other lamps or relative to other lamps.
The radiation modulation of the individual lamps or radiation sources is preferably synchronized over time for at least some of the lamps or in a fixed time correlation relative to one another, although in certain applications radiation modulations that are not synchronized over time can also be advantageous.
Pursuant to one particularly advantageous embodiment of the invention, the degree of modulation, and especially the depth of modulation, of the radiation given off by the radiation source,xe2x80x94possibly also varying from radiation source to radiation sourcexe2x80x94, is independent from the radiated light intensity. This so-called absolute modulation is thus independent from the basic level or DC signal with which the radiation source or lamp is controlled. This embodiment of the invention has the advantage that during the increase of the intensity of the radiation source, which if possible is to be undertaken rapidly, the complete control can be utilized and is not limited by too great of a modulation of its intensity.
In differently laid out applications, however, an embodiment of the invention is more advantageous where the degree or depth of modulation is dependent upon the radiated intensity of the radiation source. This so-called relative modulation, where, for example, the strength of the alternating current control signal depends upon or is proportional to the strength of the DC control signal of the radiation source, has the advantage that the relative degree of modulation is constant or varies only to a slight extent, as a result of which the detection of the modulation and the evaluation are more straight-forward and can be carried out with fewer expensive and complicated devices.
Pursuant to a further embodiment of the invention, the degree or depth of modulation is controlled or also actively regulated.
Pursuant to a further very advantageous embodiment of the invention, the lamp intensity and/or modulation itself is modulated with respect to pulse width. Pursuant to an alternative or additional embodiment of the invention, the radiation of the radiation source is modulated with a data processing program by using tabular values. A further very advantageous embodiment of the invention consists in that the radiation is modulated for the pulse width modulation by altering the register frequency of generators.
The lamp output is varied by modulating the pulse width. In this connection, the radiation intensity is a function of the filament temperature which however in the stationary, steady state corresponds directly with the lamp output.
The radiation of the radiation source is preferably modulated by means of a modulation of the control signal or signals for the radiation source or the lamps. As will be described in detail subsequently, the location at which the control signal is modulated within the signal generation can be selected as a function of the requirements and conditions. In this connection, it is particularly advantageous if the control signal, after generation thereof, is modulated immediately prior to being conducted to the radiation source or the lamps.
The present invention can be used with great advantage for determining the temperature, reflectivity and/or emissivity of an object, for example in conjunction with an apparatus for the thermal treatment of substrates, for example in a furnace in which the substrates are to be rapidly heated up and cooled with a prescribed temperature gradient that is as precise as possible.
Thus, pursuant to the invention the radiation given off by at least one radiation source, for example a heat lamp, and the radiation resulting from the object that is to be heated up, are determined, whereby the radiation from the object is a combination of the radiation emitted from the object and the radiation reflected at the object. As a result of the two measurements, it is possible to correct the radiation of the radiation sources reflected from the object and hence to determine the emitted radiation, in other words the thermal radiation of the object, that is normally, and also in the case of a wafer, no black-body radiator. By knowing the emissivity of this object, it is now possible to calculate back to the radiation of a black-body.
Pursuant to the present invention, the amplitudes of the modulated components, which are also designated as alternating current or alternating voltage (AC) components, are placed into a ratio relative to one another and the components are measured by the radiation detector provided for the object and by the radiation detector provided for the radiation sources. The number that results from the amplitude ratio is in a first approximation proportional to the reflectivity of the object, for example the wafer. This number is now used two times for the further evaluation. First of all, it is used in order to differentiate the radiation emitted from the object, in other words the thermal radiation of the object, from the radiation of the radiation source reflected at the object. Second of all, this number is used in order to scale back the radiation emitted by the object, in other words the thermal radiation, to the radiation of a black-body of the same temperature. By using the thereby obtained, scaled-back temperature value in the inverted Planckian formula, there then unequivocally results a temperature. Since the known amplitude condition of the modulations is thus used two times during the evaluation, this must be measured as precisely as possible in order to obtain precise values during the evaluation and the determination of the temperature. The inventive method enables a considerably more precise determination of this amplitude ratio since the modulation pyrometers for each heating state can be optimally prescribed and not only the modulation but also the evaluation thereof are considerably simplified.
Pursuant to a particularly preferred embodiment of the invention, the first detector is a radiation detector that in a simple and reliable manner measures the radiation given off by the radiation source. In this connection, the radiation given off by the radiation source is advantageously conducted to the radiation detector via optical lines or light channels. In order to ensure an accurate measurement, the radiation sources and the optical lines or light channels are disposed relative to one another in such a way that the first radiation detector generates a signal that is free of influences from filament holding mechanisms or other means that adversely affect radiation flux or the radiation temperature of the radiation source.
Pursuant to another embodiment of the invention, the first detector can be a temperature sensor, such as a thermo element, with which the lamp temperature, and hence the radiated intensity, can be determined.
Pursuant to a further embodiment of the invention, the first detector measures any desired parameter that is related to the radiation given off by the radiation source. Thus, for example, the intensity can be determined via an impedance measurement apparatus that measures the impedance (e.g. the ohmic resistance) of a lamp filament. By means of a suitable processing unit, it is possible, by knowing the nsity relationship of the radiation source, such as a heat lamp, to determine the intensity, or a parameter proportional thereto, given off thereby.